INDEX

Explanations

references to creatures in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

creature or creatures

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rungsseite

-0.83

ſelves

-0.77

ſelf

-0.77

 juſ

-0.74

ロウィン

-0.72

﹧

-0.70

 purpoſe

-0.70

<unused17>

-0.70

[@BOS@]

-0.70

<pad>

-0.70

POSITIVE LOGITS

 creature

0.90

creature

0.87

 Creature

0.86

Creature

0.84

 criatura

0.64

 creatures

0.64

creatures

0.55

 Creatures

0.55

 organism

0.51

 criaturas

0.47

Activations Density 0.010%

references to creatures in various contexts

creature or creatures

No Comments

No Known Activations